200 CARNEGIE INSTITUTION OF WASHINGTON. 



motion in the preceding (and following) spots of bipolar groups is now 

 opposite to its direction during the years 1908 to 1912. Thus it is as if 

 the cyclones and tornadoes north of the earth's equator were suddenly 

 to whirl clockwise instead of counter-clockwise — an event impossible 

 without profound changes in atmospheric circulation. 



In the sun, as the last report pointed out, the rotary motion of the 

 hydrogen vortices of the upper atmosphere showed no reversal of sign 

 at the sun-spot minimum. This stabiUty presumably indicates that 

 their direction of whirl is determined, not by the action of the magnetic 

 fields below them, but mainly by the conditions existing in the higher 

 atmosphere, where marked variations in solar activity have been 

 shown to be unaccompanied by fundamental changes in the rotation 

 law. Within the photosphere, however, a very different condition of 

 things must exist. In fact, the reversal of spot polarity provides a 

 possible clue to the cause of the sun-spot cycle, strongly suggesting, to 

 say the least, that its seat is within the sun, in spite of all attempts to 

 explain the cycle as a result of the influence of planetary or other 

 external phenomena. 



The inclination of the solar magnetic axis, now definitely shown to 

 make an angle of about 6° with the sun's axis of rotation, will provide 

 food for thought to students of the earth's magnetism. Two bodies, 

 the earth and the sim, rotating in the same direction, are now known 

 to be magnets of the same polarity, with magnetic axes which fail to 

 coincide with their rotation axes. The analogy, it may be hoped, will 

 serve as a useful guide to the theoretical studies of magneticians. 



The successful investigator must devote much of his attention to the 

 ehmination of sources of error and the critical investigation of plausible 

 hypotheses, even though the latter may prove in the end to be false 

 guides. The relativity hypothesis indicates that the lines of the solar 

 spectriun should be shifted very appreciably toward the red when com- 

 pared with the corresponding lines in terrestrial sources. The test is 

 rendered extremely difficult by the conditions existing in terrestrial 

 light-sources, which produce peculiar and variable shifts of the lines 

 that must be used as standards; but through persistent and painstaking 

 work these difficulties have been overcome, thus furnishing reUable 

 standards of comparison and supplying a negative answer to the rela- 

 tivity criterion. The uncertainties experienced in measuring the mem- 

 bers of closely adjacent pairs have also been reduced, and the displacing 

 effect of scattered light has been shown to be negligible under the con- 

 ditions existing on Mount Wilson. Thus the way has been prepared 

 for the study and elucidation of the most minute shifts of solar lines. 



While solar research still retains its prominent place in the work of the 

 Observatory, the development of stellar and nebular investigations 

 continues to increase. The direct photography of nebulse has not 

 only recorded new and interesting details of structure, but has also 



